CN109313272B

CN109313272B - Improved GNSS receiver using velocity integration

Info

Publication number: CN109313272B
Application number: CN201780026920.2A
Authority: CN
Inventors: F-X·马尔梅
Original assignee: Centre National dEtudes Spatiales CNES
Current assignee: Centre National dEtudes Spatiales CNES
Priority date: 2016-04-01
Filing date: 2017-03-29
Publication date: 2023-10-27
Anticipated expiration: 2037-03-29
Also published as: US10942281B2; US20190094385A1; WO2017167791A1; CN109196382A; EP3226034A1; CN109313272A; WO2017167943A1; CN109196382B; US20200309961A1

Abstract

An improved GNSS receiver is disclosed that determines the position of the receiver by combining a first position determined from standard PVT and a second position determined by integrating velocity from standard PVT. The combination is based on a weighting of the error budgets of the first location and the second location. The improved receiver is preferably based on a standard receiver with an additional software module that receives and processes data sent from the standard receiver via, for example, NMEA messages. The improved receiver allows a more accurate and smoother trajectory to be determined in a simple manner.

Description

Improved GNSS receiver using velocity integration

Technical Field

The present invention relates to GNSS receivers. More specifically, the GNSS receiver of the present invention provides improved navigation accuracy and smoothness, particularly in urban canyons, without the need to use complex multipath mitigation techniques and/or external assistance data.

Background

The use of navigation receivers is becoming more and more common in everyday life. More commonly, in-vehicle electronics, smart phones, tablet computers of automobiles include navigation receivers, and applications running thereon capture information about the location and trajectory of the end user as input.

Navigation receiver dependencyThe L-band RF signals transmitted by the middle earth orbit satellites are typically included in a constellation that includes tens of middle earth orbit satellites to cover most of the surface of the earth, such as GPS ^TM (US)、Galileo ^TM (Europe), glonass ^TM (Russian) and Beidou ^TM (China). These constellations are specified under the generic acronym of GNSS (global navigation satellite system).

The GNSS carrier signals are modulated by a pseudo-random code and navigation message that allow the calculation of the pseudo-range between the receiver and a particular satellite. With a minimum of four pseudoranges, the position, velocity and time (PVT) of the receiver can be calculated. In receivers of the type used by consumers, the location information is the location information that is directly used to calculate the navigation solution.

PVT measurements are subject to a number of errors, some of which are inherent to the measurement principle used (i.e. due to deviations in the trajectory of the RF signal through the atmosphere-ionosphere and troposphere-due to changes in the orbit of the satellites), inherent to receiver and satellite imperfections (e.g. clock bias), or inherent to some configuration of satellites in view at a certain moment (i.e. the altitude of the satellites on the horizon; low dispersion of visible satellites-high accuracy dilution or DOP). Many corrections can be used to mitigate these errors by using specific processing techniques that are only available to certain types of receivers. For example, a dual frequency receiver may mitigate ionospheric errors with gains ranging from tens of meters to several meters in accuracy, and even better when combined with accurate satellite orbits and clocks that provide accurate point positioning (PPP) -accuracy of tens of centimeters. Differential GPS and real-time kinematic solutions provide similar accuracy through integration of external information (relative positioning with respect to multiple fixed reference stations with known locations).

It is more difficult to mitigate some errors that depend on the receiver position in a consistent and efficient manner, especially when the position is surrounded by multiple objects reflecting navigation RF signals and/or masking multiple satellites that should be in line of sight (LOS) at a certain time. Under such conditions, the accuracy of the PVT calculation may be very poor, with all other error causes being equal when acquiring GNSS signals and when tracking said signals.

In urban canyons (i.e., streets between tall buildings), multipath can increase not only the errors in determining satellite pseudoranges (user equivalent range errors or UEREs), but also the errors in determining (geometric) dilution of precision (GDOP or DOP), since the field of view of the antenna will be narrower, thus limiting the increase in precision due to the use of additional satellites.

Degradation in the UERE is caused by signal impairments of the particular satellite acquired or tracked by the tracking loop. The tracking of satellites relies on the maximization of the correlation function between the acquired code signal and the plurality of local replicas generated by the receiver of each satellite-specific code signal. The correlation function will be corrupted by multipath and may not be acquired properly or the satellite may be lost. Even if signal tracking is still possible, signal impairments will affect the shape of the correlation function, degrading the pseudorange estimates and UERE.

Thus, most mitigation techniques rely on specific treatments applied at the relevant stage. For a general overview of these prior art techniques that may be applied, see, for example, bhuiyan et al, advanced Multipath Mitigation Techniques for Satellite-Based Positioning Applications, international Journal of Navigation and Observation, volume 2010,Article ID 412393. Among these techniques are: using narrow correlators (i.e., correlators spaced far less than the chip or code length); each tracking channel uses a plurality of correlators instead of a standard number of three and performs a double increment calculation; a multipath estimation delay locked loop that uses the output of the correlation function to estimate the complete navigation signal of a determined satellite, including LOS and non-LOS signals.

All these solutions are complex and expensive to implement at the hardware level or at the software level. In any case, they are currently not implemented in standard consumer receivers of the type used in smart phones or car navigation systems.

It is therefore important to find a solution for mitigating the effects of multipath or other types of impairments on navigation solutions (e.g., strong local ionospheric interference that impairs signals from one or more satellites), which can be implemented in standard receivers available to the general public.

The present invention discloses a solution to overcome the above drawbacks.

Disclosure of Invention

To this end, the invention discloses in particular a GNSS receiver configured to use a velocity measurement that is readily available whenever a condition is met, so that it is determined or possible that the velocity measurement will produce the receiver position with a better expected accuracy than using the position data directly.

To this end, the invention discloses a GNSS receiver comprising: the first calculation logic is configured to calculate a first position of the GNSS receiver based on position data in the PVT vector and a second position based on an integration result of the velocity data in the PVT vector; the second calculation logic is configured to estimate a first error budget for the first position and a second error budget for the second position, and cause the first calculation logic to output the calculated position of the GNSS receiver as a weighted average of the first position and the second position, wherein a first weight for the first position and a second weight for the second position in the calculated position are calculated by one or more of the first calculation logic or the second calculation logic based on the first error budget and the second error budget.

Advantageously, the first weight of a first one of the calculated positions is equal to the proportion of the first error budget in the sum of the first error budget and the second error budget, and the second weight of a second one of the calculated positions is equal to the proportion of the second error budget in the sum of the first error budget and the second error budget.

Advantageously, the first weight is equal to one hundred percent if the first error budget is lower than or equal to a combination of the second error budget and the first threshold.

Advantageously, the second weight is equal to one hundred percent if the second error budget is lower than a combination of the first error budget and the second threshold.

Advantageously, the second location uses one or more of the following based on decision criteria: i) The calculated position of the GNSS receiver at a time period preceding the current time period (epoch) is taken as an initial value of the integral of the velocity data in the PVT vector, or ii) the position of the GNSS receiver at a previous time period input from the positioning assistance is taken as an initial value of the integral of the velocity data in the PVT vector.

Advantageously, the positioning assistance comprises one or more of a map matching algorithm, an output of an inertial navigation system, or a Wi-Fi positioning system.

Advantageously, the estimated error budget depends on one or more of a configuration of the GNSS receiver, a geometry of GNSS satellites in view of the GNSS receiver, or a topology in an environment of the GNSS receiver.

Advantageously, the estimated error budget is estimated based on one or more of dilution of precision (DOP), user Equivalent Range Error (UERE), user Equivalent Range Rate Error (UERRE), or a probabilistic estimate of random walk along the integrated speed of its path between the previous epoch and the current epoch.

Advantageously, the first error budget is equal to DOP times UERE.

Advantageously, the second error budget is equal to the product of DOP times UERRE times the duration within which the speed is integrated.

Advantageously, the probabilistic estimate of the random walk of the integrated velocity is modeled by an ellipsoid.

Advantageously, one or more of the UERE and UERRE are a combination of one or more of a parameter value, a carrier-to-noise ratio correlation value, a high correlation value, or a signal correlation value.

Advantageously, one or more of the first error budget or the second error budget is an estimated error budget, said prediction being based on a combination of an estimated future trajectory of a GNSS receiver and a probability of said estimated future trajectory of one or more of a configuration of said GNSS receiver, a geometry of GNSS satellites in view of said GNSS receiver, or a topology in an environment of said GNSS receiver.

Advantageously, one or more of the UERE and UERRE are predefined and stored in a database accessible to the first calculation logic.

Advantageously, the GNSS receiver of the invention further comprises a third calculation logic configured to calculate heading information from the PVT vectors.

The invention also discloses a positioning method using the GNSS receiver, which comprises the following steps: calculating one or more of a first position or a second position of the GNSS receiver, the first position being based on position data in the PVT vector calculated by the GNSS receiver, the second position being based on an integration result of the velocity data in the PVT vector; estimating a first error budget for the first location and a second error budget for the second location, and causing the first calculation logic to output the calculated location of the GNSS receiver as a weighted average of the first location and the second location; wherein a first weight for a first location and a second weight for a second location in the calculated locations are calculated based on the first error budget and the second error budget.

The invention is easy to implement on any GNSS receiver. Even without modifying the receiver itself. Access to a number of data points that are generally readily accessible on a standard receiver in a well-known data format (e.g., NMEA0183 or NMEA2000 protocols) will allow the GNSS receiver to practice the present invention. The present invention improves positional accuracy in both harsh and benign environments. In so doing, it does not require external data, such as augmentation data, real-time kinematics, or information about the receiver environment topology.

Drawings

The invention and its advantages will be better understood by reading the following detailed description of specific embodiments, given purely by way of non-limiting example, with reference to the accompanying drawings, in which:

figure 1 shows the multipath problem observed from a GNSS receiver of the prior art;

FIGS. 2a and 2b illustrate typical errors of the velocity of the impairment and velocity-based position measurements in the GNSS receiver of the present invention in many embodiments of the present invention;

figures 3a and 3b show examples of parameters used in many embodiments of the invention to calculate an estimated error budget for two position and velocity measurements of a standard GNSS receiver of the invention.

Figure 4 shows a display of a prior art car navigation system;

FIG. 5 shows the architecture of a GNSS receiver modified in many embodiments of the invention to implement the invention;

figure 6 shows a flow chart of the processing steps of the calculation logic of the receiver of the invention in many embodiments of the invention;

FIG. 7 shows a first example of a comparison of the position calculations of a prior art GNSS receiver and the GNSS receiver of the present invention in various embodiments of the present invention;

fig. 8 shows a second example of a comparison of a prior art GNSS receiver and the position calculation of the GNSS receiver of the invention in various embodiments of the invention.

Detailed Description

Fig. 1 illustrates the multipath problem observed from a GNSS receiver of the prior art.

The receiver 110 is located between two buildings 120. Some of the signals 130 arrive at the receiver in a straight line. The other signals 140 arrive at the receiver after reflection from the building. Standard receivers cannot distinguish between direct signal 130 and reflected signal 140, resulting in a positioning error. The multipath error at a given location of the receiver 110 will depend on the characteristics of the obstruction at that location (altitude, roughness of its surface, etc.), but also on the altitude of the satellites in view, and thus on the time of day and on the weather conditions.

It is difficult to correct multipath errors, especially because they are location and time dependent, and therefore require a significant amount of memory and/or processing power. Furthermore, receivers configured or assisted to easily cope with other types of errors, particularly ionospheric errors, such as dual-frequency receivers, are inefficient in multipath environments, particularly when a cold start is required.

Thus, as with other types of impairments affecting the error budget of a navigation solution, finding a solution to this problem is always a problem. But especially for simple ways that can be applied to standard GNSS receivers.

Figures 2a and 2b illustrate typical errors of the velocity of the impairment and velocity-based position measurements in the GNSS receiver of the present invention in many embodiments of the present invention.

FIG. 2a shows the change in GNSS speed over time. Line 210a shows the signal boundary at 3 Standard Deviations (SDs). GNSS velocity is a direct measurement based on pseudoranges between the receiver and satellites in view, which typically has a lower error budget than the velocity output by mass-market inertial sensors, which results from acceleration integration. Furthermore, hybrid solutions between GNSS receivers and inertial navigation systems are complex to implement (in particular as retrofit solutions), and are currently mainly limited to professional receivers, in particular in the field of avionics.

Fig. 2b shows the position calculated by integrating the velocity measurement of fig. 2 a. Line 210b shows the same 3SD boundary as line 210 a. It can be seen that this location has a random walk behaviour. In practice, the comparison of the error budget should be made after integration of the speed to determine the position.

Figures 3a and 3b show examples of parameters used in the calculation of the estimated error budget for two position and velocity measurements of the standard GNSS receiver of the invention in various embodiments of the invention.

FIG. 3a is taken from

https:// en.wikipedia. Org/wiki/error_analysis_for_the_global_positioning_sys-tem m and trust its author. It shows factors affecting the position error budget, or UERE and the order of magnitude of these errors. It can be seen that the larger factors are ionospheric effects, with possible errors of 10m (-5/+5m), followed by code tracking jitter (-3/+3m), satellite ephemeris, satellite clock, multipath distortion, tropospheric effects and carrier phase.

FIG. 3b was taken from Precise Velocity and Acceleration Determination Using a Standalone GPS Receiver in Real Time, J.ZHANG, doctor paper (Table 6.2) and trusted its authors. It shows factors affecting the speed error budget, or User Equivalent Range Rate Error (UERRE) and the order of magnitude of these errors. We can see that in general the error budget for speed is two orders of magnitude lower than UERE. This is mainly due to the fact that the velocity is measured from the doppler effect. Moreover, even for significant errors, their variation is much lower. For example, the ionospheric effect should drop from an error of up to 10 meters to an error magnitude of 1 cm/s. Multipath errors become negligible.

In accordance with the present invention, the velocity is then integrated once to create a new calculated position, which typically has a lower error budget.

There is a condition that the error of the integrated speed may become larger than the error of the position. This may be the case, for example, when the product of UERE x DOP becomes lower than the standard deviation of the integrated speed, especially when the integration time is long. It is therefore very advantageous to be able to then switch from a speed measurement to a position measurement. It is also advantageous to use the exact position as the initial position of the integration position.

Fig. 4 shows a display of a prior art car navigation system.

FIG. 4 is a slave

http:// www.gpspower.net/garmin-tutorials/141052-secret-start-commands-hide den-menu-garmin-devices.

The National Marine Electronics Association (NMEA) has defined a standard sentence in a format for outputting from a GNSS receiver a large amount of information that can be used to implement the present invention using a standard receiver. Two criteria apply (NMEA 0183 and NMEA 2000).

The receiver displays (and outputs at the port) in particular: position 410, velocity 420, C/N0 430, which is specified by a number that is characteristic of its constellation, DOP 450, and altitude 460 for each satellite in view. Other data may be obtained, such as azimuth, raw pseudorange measurements, estimated PVT residuals, and the like.

FIG. 5 illustrates the architecture of a GNSS receiver modified to implement the present invention in many embodiments of the present invention.

The receiver 510 is a core GNSS receiver according to the present invention having one or more antennas 511, a plurality of radio frequency processing channels 512 and digital processing capabilities 513. The exact composition of the core receiver may vary depending on the use case (military or professional applications, vehicle navigation, pedestrian specific devices or smart phones). It may be capable of processing signals from multiple constellations (GPS, glonass, beidou, galileo, etc.), and for each constellation, one, two, three or more frequencies. The core receiver may be implemented in an FPGA or ASIC, or as a software defined receiver, in multiple chips or in a single chipset.

As described above, the receiver 510 may output a message 531,532 in NMEA format (or in other formats, including proprietary formats defined by the manufacturer). Typically, the manufacturer derives its message in its format or NMEA format. Most of them publish the structure of the messages they output. There is enough information to implement the invention as long as the outgoing message includes the PVT calculated by the receiver. The outputs 531 and 532 have been separated from what is mentioned below, but this can be done physically through a single interface.

These messages are fed to the calculation logic 520, which calculation logic 520 will calculate the navigation solution according to the invention, i.e. the position according to which the weighted average of the first position is the calculated PVT, and the second position which is the velocity of the calculated PVT. The processing logic 520 may be implemented in a hardware add-on module or in a software module loaded in or added to a memory included in the receiver. The software may also be available as an application downloaded and/or installed on the smart phone. It may also be fully integrated with the core GNSS receiver 510 on the same board or in the same FPGA or the same ASIC.

Parameters 521 are input from the receiver and possibly from a table stored in a memory contained in the processing logic 520 or adjacent to the processing logic 520.

The parameters input from the core receiver through port 531 may include some or all of those listed in the description of fig. 4 as well as other parameters that may be found useful for mixing the first and second locations, as further described below with respect to fig. 6.

The table may include parameters such as a calculated error budget, a first error budget for a GNSS calculated position (UERE) and a second error budget for a GNSS calculated velocity (UERRE), or an error budget for a position calculated by integrating the velocity. These parameters may be preset values for UERE and UERRE. They may also include preset values for UERE and UERRE, which depend on the altitude of the satellites and/or their current C/N0 and/or GNSS signal type. They may also include the configuration of the receiver (type or model of antenna, type or model of RF processing stage, type or model of digital processing stage). They may also include a table of multipath effects at locations on the receiver's trajectory. They may also include different threshold values that will be used to calculate the weight of a first location calculated from P in PVT, and a second location calculated from V in PVT, as provided by the present invention. For example, for a pedestrian looking for only coarse positioning, the maximum value of UERE may be set to 50m, such that P measurements are discarded entirely when the actual value is higher. Also, the maximum value of the UERRE is 1m/s. For users requiring higher accuracy, the threshold will be lowered.

The first calculation logic 522 extracts P and V from PVT data output from the core receiver through a port or interface 532. P is the first position and the second position is calculated by integrating V. In an embodiment, the distance calculated from the speed integral will be added to the position previously calculated by the first calculation logic 522 to calculate the second position. However, some variations are possible, for example, the second location 540 may be entered from an external source based on decision criteria, such as a threshold for an error budget or determining whether the receiver is considered to be located in an area with sunny days. The external source may be an inertial module collocated with the GNSS receiver, such as a device including one or more of an accelerometer, a gyroscope, and a magnetometer, as is available in a smart phone. The external source may also be the output of a map matching algorithm, as is available in a car navigation system, or any other source of location of the user or device.

The second calculation logic 523 calculates the error budget over time from data that may combine the predetermined information from the parameter inputs with information from the GNSS receiver, such as the configuration of the receiver (antenna type or model, type or model of RF processing stage, type or model of digital processing stage), in view of the geometry of the GNSS satellites of the GNSS receiver (provided by NMEA messages), or topology in the environment of the GNSS receiver, depending on how the received signals are multipath affected according to the multipath effect map available in the table. The second calculation logic may also use a C/N0 value for a channel of a satellite in the processing view, the value provided by the NMEA message. It may also use the DOP also provided by the NMEA message. It may also use the probability of a random walk of the integrated velocity. In an embodiment, the random walk of the integrated velocity may be modeled by an ellipsoid. In an embodiment, the first error budget is calculated as DOP times UERE. In an embodiment, the second error budget is calculated as the product of DOP times UERRE times the duration within which the speed is integrated.

One or more of the first calculation logic 522 or the second calculation logic 523 is configured to calculate a first weight of the first error budget and a second weight of the second error budget, e.g., as a percentage of the total error budget. The computation may be divided between the first computation logic and the second computation logic to optimize the computation without departing from the scope of the claimed invention. The weights may be calculated as a proportion of each error budget in the sum of the first and second error budgets.

In an embodiment, the first weight is equal to one hundred percent if the first error budget is lower than or equal to a combination of the second error budget and the first threshold. The first threshold may be static and extracted 521 from a table of parameter values. The first threshold may also be dynamically modified based on the reception conditions at the GNSS receiver extracted from the NMEA message.

In an embodiment, the second weight is equal to one hundred percent if the second error budget is lower than a combination of the first error budget and the second threshold. The second threshold may be static and extracted 521 from a table of parameter values. The second threshold may also be dynamically modified based on the reception conditions at the GNSS receiver extracted from the NMEA message.

According to the invention, the first calculation logic is then caused to output the position of the GNSS receiver as a weighted average of the first position and the second position. The output location is a navigation solution. In some embodiments, only the first position or only the second position is considered to output the position of the GNSS receiver based on the relative values of the first and second error budgets.

It should be noted that even though this data may improve the navigation solution, good results may be obtained without external data. This is due to the fact that the preset UERE and UERRE can be adjusted using parameters (e.g. DOP, C/N0, altitude of satellite) readily available from the GNSS receiver via interfaces 531,532 to calculate the first and second error budget adapted to the actual conditions of the GNSS signal or reception.

Fig. 6 shows a flow chart of the processing steps of the computational logic of the receiver of the present invention in various embodiments of the present invention.

Step 611 is performed in first calculation logic 522, in accordance with the present invention.

In step 6111, a time reference t is obtained from the GNSS receiver. At cold start, the time reference t will be a period after which a first P value with a confidence interval better than the threshold is obtained.

In step 6112, P is obtained from the PVT calculated in the digital processing stage of the GNSS receiver.

At the same time reference, V is obtained from the PVT calculated by the digital processing stage of the GNSS receiver in step 6113. In step 6116 v is integrated once.

Thus, the improved position can be calculated as:

pinit is P or P 'at the start time of the improved position calculation, or the average position obtained from several P or P' values over time.

In a variant of the invention, the position P' is obtained from an external source in step 6115, as explained above with respect to fig. 5. In this variant, logic may be included in the first calculation logic to implement decision criteria for selecting P or P' at time t. The decision criteria may take into account the accuracy of the GNSS location sent by the NMEA message.

Then, in steps 6117 and 6118, respectively, the first position and the second position are calculated.

In step 6119, the first position, the second position and the weights calculated in the first calculation logic (embodiment not shown in the figure) or in the second calculation logic (as shown in the figure) will be considered to calculate the position to be output as the navigation solution.

Step 612 is performed in the second calculation logic 523 in accordance with the present invention.

In step 6121, the parameter values are obtained from the parameter inputs 521, either through the interface 531 to the GNSS core receiver or through the external input 540.

In steps 6122 and 6123, a first error budget and a second error budget are calculated, respectively, as explained above in connection with fig. 5.

In step 6124, the weight of the error budget is calculated for transfer to the first calculation logic. Alternatively, the error budget may be transferred to the first calculation logic so that it calculates the weights.

Fig. 7 shows a first example of a comparison of position calculations of a prior art GNSS receiver and a GNSS receiver of the present invention in various embodiments of the present invention.

The situation shown in fig. 7 is a trajectory of a motor vehicle in a suburban area, where the density of buildings is rather low. A receiver (GPS, glonass, galileo, beidou) acquires and tracks four constellations. The duration of the trace is about 50 minutes. Simple assumptions are made about the error budget: the UERE is set to a fixed value of 1m and the UERRE is set to a fixed value of 0.1m/s. These assumptions are optimistic for the UERE and pessimistic for the UERRE, considering the local environment. In this case, no dynamic modification of the error budget is added. In this example, a simple choice of P or V for integration is made according to a lower error budget:

-if the first error budget is lower than the second error budget, selecting the first position calculated from P in step 6117 as a navigation solution;

if the second error budget is lower than the first error budget, then the second position calculated in step 6118 is selected (as the previous first position to which the distance calculated by the integral of V was added).

In fig. 7, the trajectory of the vehicle calculated by the known unassisted GNSS receiver is represented by line 710. The trajectory of the vehicle calculated by the GNSS receiver to which the device 520 configured according to the invention is added is represented by the line 720.

It can be seen that line 710 shows a number of errors 730, 740, 750 in the trajectory, which do not follow the road. This illustrates some of the advantages that are brought to the prior art solutions by the simple implementation of the invention in some embodiments of the invention.

FIG. 8 illustrates a second example of a comparison of position calculations of a prior art GNSS receiver and a GNSS receiver of the present invention in various embodiments of the present invention.

The situation shown in fig. 8 is a trajectory of a motor vehicle in a city center, where the density of buildings is quite high. The receiver and the GNSS constellation acquired and tracked are used as in the case of fig. 7. In this case, the duration of the trace is about 1 hour and 45 minutes. Simple assumptions are made about the error budget: the UERE is set to a fixed value of 2m and the UERRE is set to a fixed value of 0.2m/s. These assumptions are optimistic for the UERE and pessimistic for the UERRE. Other conditions in the case of fig. 8 are the same as those of fig. 7.

In fig. 8, the vehicle trajectory calculated by the prior art unassisted GNSS receiver is represented by line 810. The vehicle trajectory calculated by the GNSS with the addition of the device 520 configured according to the invention is represented by the line 820.

It can be seen that line 810 shows a plurality of errors 830, 840, 850 in the trajectory that do not follow the road. In some embodiments of the invention, the use case further illustrates some advantages of the invention over prior art solutions in a constrained environment by the implementation of the invention.

Notably, the present invention not only provides a more accurate navigation solution, but also provides a smoother trajectory because it is immune to errors 830, 840, and 850.

The improvements described and claimed in some embodiments of the invention will still improve the navigation solution compared to the above case. For example, an integration of dynamic parameters representing the reception conditions at the GNSS receiver, or a weighting calculated based on two positions of its error budget, or an integration of external navigation aids. It is worth noting that these improvements can be obtained by means of a standard GNSS receiver, to which the present invention can be added in a simple manner.

Standard receivers are for example (available from

http:// gpswworld.com/wp-content/uploads/2015/01/gpsword_ 2015ReceiverSurv ey.pdf) gpsword is marked in the Receiver survey2015 database as a standard Receiver for entertainment, hand-held, navigation, vehicle tracking, or board for OEM applications.

The invention is described above in the context of a receiver using GNSS signals. It can also be applied to receivers using signals transmitted by pseudolites, e.g. by location ^TM Pseudolites for sale.

The vector V used to calculate the second position may also be used to determine a navigational heading, either alone or in combination with outputs from other heading sensors (e.g., magnetometers), or a combination of accelerometers and gyroscopes, or a map matching algorithm, in accordance with the present invention. The navigational heading may be used by an application, for example, to give a direction of a point of interest relative to the navigational heading. It can also be used to calibrate other heading sensors. This may be applied to receivers for pedestrian use or to receivers located on manned or unmanned cars, aircraft or boats.

The examples disclosed in this specification merely illustrate some embodiments of the invention. They do not in any way limit the scope of the invention as defined by the appended claims.

Claims

1. A GNSS receiver configured to calculate PVT vectors, the GNSS receiver comprising:

-first calculation logic configured to calculate a first position and a second position of the GNSS receiver, the first position being based on position data in the PVT vector, the second position being based on a result of integration of velocity data in the PVT vector;

second calculation logic configured to estimate a first error budget for the first position and a second error budget for the second position, and to cause the first calculation logic to output the calculated position of the GNSS receiver as a weighted average of the first position and the second position, the first error budget being based on a preset value of a user equivalent range error UERE, and the second error budget being based on a preset value of a user equivalent range rate error UERRE,

wherein a first weight for the first location and a second weight for the second location in the calculated locations are calculated by one or more of the first calculation logic or the second calculation logic based on the first error budget and the second error budget.

2. The GNSS receiver of claim 1 wherein a first weight of the first one of the calculated positions is equal to a proportion of the first error budget in a sum of the first error budget and the second error budget and a second weight of the second one of the calculated positions is equal to a proportion of the second error budget in a sum of the first error budget and the second error budget.

3. The GNSS receiver of claim 1, wherein only the first position or only the second position is considered to output the calculated position of the GNSS receiver based on relative values of the first and second error budgets.

4. The GNSS receiver of claim 1, wherein the second location uses one or more of the following based on decision criteria: i) The calculated position of the GNSS receiver at a time period prior to the current time period is taken as an initial value of the integral of the velocity data in the PVT vector, or ii) the position of the GNSS receiver at a previous time period input from a positioning assistance device is taken as an initial value of the integral of the velocity data in the PVT vector.

5. The GNSS receiver of claim 4, wherein the positioning assistance device comprises one or more of a map matching algorithm, an output of an inertial navigation system, or a Wi-Fi positioning system.

6. The GNSS receiver of claim 5 wherein the estimated error budget is dependent on one or more of a configuration of the GNSS receiver, a geometry of GNSS satellites in view of the GNSS receiver, or a topology in an environment of the GNSS receiver.

7. The GNSS receiver of claim 5, wherein the first error budget and the second error budget are further calculated based on one or more of a dilution of precision DOP and a probability estimate of a random walk along the integrated speed of its path between a previous epoch and a current epoch.

8. The GNSS receiver of claim 7, wherein the first error budget is equal to the DOP times the UERE.

9. The GNSS receiver of claim 7, wherein the second error budget is equal to a product of the DOP times the UERRE times a duration within which the velocity is integrated.

10. The GNSS receiver of claim 9 wherein the probability estimate of the random walk of the integrated velocity is modeled by an ellipsoid.

11. The GNSS receiver of one of claims 7 to 10, wherein one or more of the UERE and the UERRE are further calculated based on one or more of a parameter value, a carrier-to-noise ratio correlation value, a high correlation value, and a signal correlation value.

12. The GNSS receiver of one of claims 7 to 10, wherein one or more of the first error budget or the second error budget is a predicted estimated error budget, the prediction being based on a combination of: an estimated future trajectory of the GNSS receiver, and a probability of the estimated future trajectory of one or more of a configuration of the GNSS receiver, a geometry of GNSS satellites in view of the GNSS receiver, or a topology in an environment of the GNSS receiver.

13. The GNSS receiver of one of claims 7 to 10, wherein one or more of the UERE and the preset value of the UERRE are stored in a database accessible to the first calculation logic.

14. The GNSS receiver of one of claims 1 to 10, further comprising third calculation logic configured to calculate heading information from the PVT vectors.

15. A positioning method using a GNSS receiver, comprising:

-calculating a first position and a second position of the GNSS receiver, the first position being based on position data in a PVT vector calculated by the GNSS receiver, the second position being based on the result of integration of velocity data in the PVT vector;

-estimating a first error budget for said first location and a second error budget for said second location, and

calculating a position of the GNSS receiver as a weighted average of the first position and the second position, the first error budget being based on a preset value of a user equivalent range error UERE and the second error budget being based on a preset value of a user equivalent range rate error UERRE;

wherein a first weight of the first location and a second weight of the second location in the calculated locations are calculated based on the first error budget and the second error budget.